A FAIR intensive longitudinal data archive on prolonged grief in daily life

ABSTRACT

Introduction: The majority of grief research has assessed Prolonged Grief (PG) symptoms using cross-sectional surveys, which are limited in capturing within-person changes in daily grief reactions. These dynamic PG reactions can be more effectively assessed using Experience Sampling Methodology (ESM). Accordingly, this data note presents a Findable, Accessible, Interoperable, and Reusable (FAIR) archive comprising three existing ESM projects investigating PG reactions in daily life.

Methods: Participants in this archive completed three data collection phases: baseline measures (including sociodemographic and loss characteristics, and psychopathology measures), a 14-day ESM phase (rating PG reactions and contextual factors up to five times daily), and follow-up psychopathology assessments. The participants provided explicit digital consent for the use and reuse of their data in scientific research.

Results: In total, 315 people are included in this archive, with a total of 22,050 ESM-measurement points. The majority of the participants were middle-aged, identified as women, and had completed higher education. Almost half of the sample (48.73%) experienced the loss of a partner or child, the majority of these losses were due to natural causes (70.06%), and happened less than a year ago (59.24%). The data are stored in a trusted repository.

Discussion: This archive demonstrates that it is feasible to develop a FAIR archive including existing data on PG reactions in daily life. Given the resource-intensive nature and richness of these data, we encourage researchers to reuse and/or share ESM-data through this archive, helping to deepen our understanding of grief in natural settings.

HIGHLIGHTS

• This FAIR archive includes experience sampling data on prolonged grief.

• It includes data from 315 bereaved people and 22,050 measurement points.

• This reusable resource helps to advance our understanding of grief in daily life.

Grief is a natural response to the death of a close person. While the majority of people recover within about one year after the loss (Pociunaite et al., 2023); approximately 3% develop severe grief symptoms that result in considerable distress and impairment in daily functioning (Rosner et al., 2021). In the text revision of the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR) these severe and disabling grief symptoms are defined as a Prolonged Grief Disorder (PGD) (American Psychiatric Association, 2022). PGD is characterized by yearning for, and preoccupation with, the deceased, and eight additional symptoms namely the feeling that a part of oneself died, perceiving the loss as unreal, avoidance, intense emotional pain (sadness, anger), difficulty moving on, numbness, meaninglessness of life, and loneliness.

The prevailing methodology in grief research is cross-sectional survey research that examines differences between people in terms of severity of PG symptoms. This cross-sectional research often employs retrospective self-report measures, wherein people rate their PG symptom severity in the past month. While anecdotal and theoretical work has proposed that grief can be a dynamic and fluctuating process (Arizmendi & O’Connor, 2015; Stroebe & Schut, 1999), cross-sectional research inherently lacks the capacity to examine potential changes that happen within a person. In response to this, the data presented in this data note have been collected with the intention of capturing dynamic grief experiences by investigating PG reactions in daily life.

PG reactions in daily life can be assessed using Experience Sampling Methodology (ESM), also referred to as Ecological Momentary Assessment (EMA). ESM entails capturing participants’ experiences as they occur in their natural environment, typically through the use of smartphone applications (Fritz et al., 2024). In contrast to retrospective cross-sectional studies, during ESM people report on their experiences multiple times a day (e.g. five times), over an extended period of time (e.g. 14 days). The use of ESM is advantageous as it can capture time- and context-dependent changes, disentangle within-person from between-person effects, reduce recall bias, and produce a more ecologically valid representation of grief experiences (Fritz et al., 2024; Hruska et al., 2025; Myin-Germeys et al., 2024). ESM research has shown that assessing PG reactions in daily life using ESM is feasible and acceptable and is not resulting in reactivity effects (Franzen & Lenferink, 2024; Lenferink et al., 2022b; Mintz et al., 2024). Additionally, some PG reactions were found to fluctuate more than others (Ergun et al., 2025; Lenferink et al., 2024), which seem to be related to contextual factors (Pociunaite et al., 2024).

In this data note, we present a Findable, Accessible, Interoperable, and Reusable (FAIR) archive that includes three existing datasets from three different ESM projects investigating PG reactions¹ in daily life. The FAIR guiding principles for scientific data were first introduced in 2016 as part of the larger movement towards greater openness and transparency in science across all disciplines (Wilkinson et al., 2016). The FAIR guiding principles have the potential to accelerate research transparency and scientific discovery, while also addressing the high burden on participants and researchers (Hruska et al., 2025; Prakash et al., 2023; Sadeh et al., 2023). Reducing the high burden is of particular relevance in vulnerable and under-represented samples, such as ESM-research among bereaved people. Moreover, combining multiple projects creates opportunities for novel research findings that would otherwise be impossible in single studies (Kassam-Adams et al., 2020). In general, data reuse allows researchers to study the etiology and trajectories of mental health outcomes across populations, to identify risk and protective factors with greater statistical power, and to conduct subgroup analyses to understand for whom, how, why, and under what conditions mental health treatments are most effective (De Haan et al., 2024; Pociunaite et al., 2023; Sadeh et al., 2023; van Tuijl et al., 2023).

Despite the generally positive outlook of trauma researchers towards data sharing and reuse, the perceived barriers remain (e.g. too much effort required for data sharing and reuse) (Prakash et al., 2023). In response, we present a data note that illustrates our approach to complying with the FAIR principles and addressing the perceived barriers with the aim of providing guidance to other researchers in their efforts to make their research FAIR. We present where our novel archive on PG reactions in daily life can be found (https://doi.org/10.17026/SS/LNFAXI), we describe the conditions under which other researchers may access the archive, we share the associated files in a variety of interoperable formats that are open access, and finally we present a detailed description of our methods and data structure for the ease of reuse.

1. Materials and methods

1.1. Data collection

Data from three projects were harmonized into this unique archive called Grief in Daily life Archive (Grief-ID Archive) available at https://doi.org/10.17026/SS/LNFAXI. The methodologies employed in the three projects were largely similar as outlined in Table 1. All people included in this archive participated in three distinct phases of data collection.

Table 1.

Similarities and differences in study design between three projects included in the archive.

	Project 1: ESM 1	Project 2: ESM 2	Project 3: ESM 3
	https://doi.org/10.17026/SS/INSTAE	https://doi.org/10.17026/SS/RRJ6OW	https://doi.org/10.17026/SS/TSUSOD
Similarities between projects
Country of data collection	The Netherlands (NL)
Experience sampling methodology (ESM)	Yes
Pre-ESM measurement (T1)	Yes
Post-ESM measurement (T2)	Yes
ESM tool	Avicenna (Ethica)
Total number of measurement occasions	70a
Measurements per day	5
Duration of ESM phase (in days)	14
Beep time stamps	8:30 – 9:30 AM, 11:30 AM – 12:30 PM, 2:30 – 3:30 PM, 5:30 – 6:30 PM, and 8:30 – 9:30 PM
Differences between projects
Date of data collection	January – March 2022	February – August 2023	April – July 2024
Recruitment outlet	Convenience sampling (social network platforms, snowballing method)	Grief treatment website www.rouwbehandeling.nl	Grief treatment website www.rouwbehandeling.nl
Language	German, Dutch	Dutch	Dutch
T1 and T2 data collection type	Clinical interview via telephone	Clinical interview via telephone	Online survey via Qualtrics
Time since loss inclusion criterion	≥ 3 months	3–6 months postloss	≥ 12 months
Exclusion criteria	High suicidal risk; Psychotic diagnosis	High suicidal risk; Psychotic diagnosis	Current suicidal ideations; Psychotic diagnosis
Cause of death	All types	All types	Traumatic loss (i.e. suicide, homicide, accident)
Associated publications	Lenferink et al. (2022b) Lenferink et al. (2024) Pociunaite et al. (2024)	Franzen and Lenferink (2024) Ergun et al. (2025)	Specker et al. (2025)

Note. ^aDay 14 is missing for some participants due to a technical error (i.e. some participants received a wrong number of notifications). ESM = Experience sampling methodology.

First, the initial baseline measures included questions assessing sociodemographic and loss characteristics, measures of psychopathology, and other measures as outlined in Table 2. People aged 18 years and above were eligible to participate. Those diagnosed with a psychotic disorder and exhibiting suicidal ideations, assessed at baseline, were excluded from further participation.

Table 2.

Overview of instruments used in each project that is part of this archive.

	Project 1: ESM 1	Project 2: ESM 2	Project 3: ESM 3
Measures in pre-ESM (T1) and post-ESM (T2)
PGD measure (T1, T2)	TGI-CA	TGI-CA	TGI-SR+
PTSD measure (T1, T2)	PCL-5
Depression measure (T1, T2)	PHQ-9
Functioning (T1)	WSAS
Self-reflection (T1)	No	SRIS	No
Rumination (T1)	No	No	RRS
Reaction to research participation (T2)	RRPQ
ESM measures
ESM-PG reactions	11 items based on Lenferink et al. (2022b)
ESM contextual factors	Location; Activity; Quality of activitya; With other people/alone; Quality of social contact; Relationship to other person.	Location; Activity; Quality of activity; With other people/alone; Quality of social contact; Relationship to other person.	Location; Activity; Quality of activity; With other people/alone; Quality of social contact; Quality of being alone; Relationship to other person.
ESM grief overall (1 item)	No	No	Yes
ESM blame (self/others)	No	No	Yes
ESM rumination	No	No	Yes

Note. ^aActivity and Quality of activity were not assessed in German speaking part of the sample. ESM = Experience sampling methodology; PCL-5 = PTSD-Checklist for DSM-5; PGD = Prolonged grief disorder; PHQ-9 = Patient Health Questionnaire; PTSD = Posttraumatic stress disorder; RRS = Ruminative Response Scale; RRPQ = Reactions to Research Participation Questionnaire; SRIS = Self-Reflection and Insight Scale; TGI-CA = Traumatic Grief Inventory-Clinician Administered; TGI-SR + =  Traumatic Grief Inventory-Self Report; WSAS = Work and Social Adjustment Scale.

Second, the ESM phase was conducted using the Avicenna application (Avicenna Research, 2024). In this phase, participants rated the intensity of their PG reactions and contextual factors five times a day, for a period of 14 days (design analog to Schoevers et al., 2021). Participants received a beep on their phone semi-randomly at 8:30–9:30 AM, 11:30 AM–12:30 PM, 2:30–3:30 PM, 5:30–6:30 PM, and 8:30–9:30 PM. If participants had not responded within 10 or 20 min, reminders were sent. Participants had 60 min to complete the ESM items after the first beep. Prior to the ESM phase, participants were provided with a video tutorial outlining the steps for installing the application and enabling notifications. If any technical assistance was needed, research assistants were available to provide it.

In the final phase, psychopathology was assessed after the ESM phase in a manner consistent with the baseline phase. For projects 2 and 3, these data were collected via telephone interviews, whereas for project 3, data were collected via online surveys. The data were collected in the Netherlands between January 2022 and July 2024.

There are several notable differences across the three projects, as denoted in Table 1. For project 1, the participants were recruited through community sampling, snowballing techniques, and social network platforms. For projects 2 and 3, participants were recruited via a public website (www.rouwbehandeling.nl, English: Grief Treatment). This website is a Dutch-language resource designed for grieving persons. This website offers information about grief, prolonged grief, and available support. People can also complete a self-report measure of PGD (i.e. the Traumatic Grief Inventory-Self Report Plus (TGI-SR+; Lenferink et al., 2022a)), and indicate whether they wish to be contacted regarding future research. If agreed, they are asked to share their contact information. This contact information was used for inviting people to participate in projects 2 and 3.

Another notable difference across the three projects is the inclusion criteria, which vary in terms of type of loss and the time since loss. Project 1 included participants who had experienced the loss of a loved one at least three months prior and were bereaved due to any type of death. Project 2 included participants whose loved one died three to six months earlier due to any type of death. Project 3 included participants who had experienced the loss at least 12 months prior and were bereaved due to a traumatic loss (i.e. accident, homicide, suicide) and experienced PGD intensity that exceeded the clinical cut-off (≥ 71; Lenferink et al., 2022a) at the time of completing the online self-report measure of PGD at www.rouwbehandeling.nl.

1.2. Ethical considerations

In accordance with the General Data Protection Regulation (GDPR), participants gave active digital consent for the use of the data for scientific purposes (i.e. to meet the research aims for the initial project) outlined in the information letter as well as for sharing the data with, and reusing the data by, other researchers for other research purposes beyond the scope of the initial project. In projects 1 and 2, five participants out of 245 did not agree to their data being shared and reused in other research projects, these participants were excluded from the archive. Supplementary Material A includes informed consent and information letter excerpts on data sharing from all three projects. The three projects included in this archive have been approved by a local institution review board (IDs: 211101; 221328; 240186). The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. The data are stored in Data Archiving and Networked Services (DANS) repository that is certified with the CoreTrustSeal – a recognized certification for trusted data repositories that meet international standards for data preservation, accessibility, and sustainability.

1.3. Instruments

Each project included a variety of questionnaires, with some substantial overlap. Table 2 provides an overview of the instruments used in each project, outlining the similarities and differences.

1.4. Retrospective measures

1.4.1. Sociodemographic characteristics

Sociodemographic characteristics were assessed with questions about participant’s date of birth (age), gender (1 (male), 2 (female), 3 (other)), and highest obtained level of education (1 (primary school), 2 (high school), 3 (vocational education), 4 (applied) university)). Several data de-identification measures were taken. First, age was recoded into three age groups: young, middle-aged, and older adults. Second, participants’ education was recoded into two broader categories, with (applied) university coded as higher education and primary, high school, and vocational education coded as other.

1.4.2. Loss-related characteristics

Questions related to the loss consisted of time since loss (in months); the relationship to the deceased (losing a 1 (partner), 2 (child), 3 (parent), 4 (sibling), 5 (grandparent), 6 (grandchild), 7 (friend), 8 (other, namely)); cause of death (1 (physical illness), 2 (accident), 3 (suicide), 4 (homicide/manslaughter), 5 (other, namely)); to what extent this loss was unexpected (1 (completely expected) to 5 (completely unexpected)); if participants ever received psychological support prior to the loss (1 (no), 2 (yes)); if participants ever received professional grief support related to the loss (1 (no), 2 (yes)). In case a participant has experienced multiple losses, they were asked to answer all questions about the loss they considered to be the most significant loss. To ensure data de-identification, the information was recoded into broader categories. In terms of the relationship to the deceased, we dichotomized scores into the loss of a partner or child, as opposed to other types of relationships. This is because the loss of a partner or child is a known risk factor for PGD (Buur et al., 2024). In terms of cause of death, we dichotomized scores into losses under traumatic circumstances (e.g. suicide, homicide, or accidents) and other causes.

1.4.3. Prolonged grief (PG) severity

PG symptoms were assessed through the Traumatic Grief Inventory-Self Report (TGI-SR+) (Lenferink et al., 2022a) and Traumatic Grief Inventory-Clinician Administered (TGI-CA) (Lenferink et al., 2023), available for free at: https://osf.io/rqn5k/ and https://osf.io/a6hmc/, respectively. TGI-CA (Lenferink et al., 2023) is an interview version of the 22-item TGI-SR + (Lenferink et al., 2022a). Using 5-point Likert scales (1 = never, 2 = rarely, 3 = sometimes, 4 = frequently, 5 = always), participants were asked to what extent they experienced each PGD reaction during the past month (e.g. ‘In the past month, did you feel alone or detached from others?’). For project 1 and 2, but not 3, the time frame was adjusted from ‘past month’ to ‘past two weeks’ to fit the ESM period. This measure assesses multiple criteria-sets for disturbed grief (i.e. DSM-5 persistent complex bereavement disorder, DSM-5-TR PGD, and ICD-11 PGD) and has good psychometric properties (Lenferink et al., 2022a; Lenferink et al., 2023).

1.4.4. Posttraumatic stress disorder (PTSD) severity

PTSD symptoms were assessed with the PTSD-Checklist for DSM-5 (PCL-5; Blevins et al., 2015; Boeschoten et al., 2018). The PCL-5 is a self-report measure that consists of 20 items. Using 5-point Likert scales (0 = not at all, 1 = a little bit, 2 = moderately, 3 = quite a bit, 4 = extremely), participants were asked to what extent they bothered by each PTSD symptom (e.g. ‘In the past month, were you suddenly feeling or acting as if the stressful experience were actually happening again (as if you were actually back there reliving it)?’). For projects 1 and 2, but not 3, the time frame was adjusted from ‘past month’ to ‘past two weeks’ to fit the ESM period. The instructions referred to ‘death of a loved one’ instead of ‘stressful event’. The PCL-5 has good psychometric properties (Blevins et al., 2015).

1.4.5. Depression severity

Depression symptoms were assessed with the Patient Health Questionnaire (PHQ-9; Kroenke et al., 2001). The PHQ-9 consists of 9 items. Using 4-point Likert scales (0 = not at all, 1 = several days, 2 = more than half the days, to 3 = nearly every day), participants indicated how often they were bothered by each of the nine symptoms during the past two weeks (e.g. ‘How often have you been bothered by having little interest or pleasure in doing things?’). The psychometric properties of PHQ-9 are good (Kroenke et al., 2001).

1.4.6. Functional impairment

Functional impairment was assessed with the Work and Social Adjustment Scale (WSAS; Mundt et al., 2002). This 5-item measure consists of 9-point Likert scales (from 0 = not at all to 8 = very severely), where participants report their difficulties in daily functioning over the past week (e.g. ‘Please indicate the extent to which the death of your loved one has generally hindered you in ability to work). The WSAS has sound psychometric properties (Mataix-Cols et al., 2005; Mundt et al., 2002; Pedersen et al., 2017).

1.4.7. Self-reflection and insight

Self-reflection and insight was assessed with the Self-Reflection and Insight Scale (SRIS; Grant et al., 2002; Sauter et al., 2010). This 20-item measure assesses self-reflection and self-insight. Using 6-point Likert scales (1 = disagree strongly, 2 = disagree, 3 = disagree slightly, 4 = agree slightly, 5 = agree, 6 = agree strongly), participants were asked to rate the extent of their self-consciousness (e.g. Reflection: ‘I frequently take time to reflect on my thoughts’, Insight: ‘I usually know why I feel the way I do’). The SRIS has good psychometric properties (Grant et al., 2002; Silvia, 2022).

1.4.8. Rumination

Ruminative Response Scale consists of reflective pondering and brooding subscales (RRS; Raes, 2009; Treynor et al., 2003). We used the 5-item brooding subscale only. Using 4-point Likert scales (1 = almost never, 2 = sometimes, 3 = often, 4 = almost always), participant were asked to indicate the frequency of their ruminative thinking tendencies in general. Brooding reflects a passive comparison of one’s current situation with something that they have not achieved (e.g. How often you think about a recent situation, wishing it had gone better?). The RRS psychometric properties are good (Treynor et al., 2003).

1.4.9. Reaction to research participation

Reaction to participating in ESM research was assessed using the Reactions to research participation questionnaire (RRPQ; Newman et al., 2001). This self-report measure consists of a 4-item Personal Benefits subscale and a 4-item Emotional Reactions subscale. Using 5-point Likert scales (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, to 5 = strongly agree), participants rated the acceptability of self-monitoring PG reactions in daily life (e.g. Personal benefits: ‘I found participating in daily diary measures in the app beneficial to me.’; Emotional reactions: e.g. ‘Participating in daily diary measures in the app raised emotional issues for me that I had not expected.’). Following prior research (Waterman et al., 2021), we referred to ‘participating in daily diary measures in the app’ instead of ‘study’ and statements were rephrased as questions. The RRPQ has good psychometric properties (Kassam-Adams & Newman, 2002; Newman et al., 2001; Waterman et al., 2021).

1.5. ESM measures

1.5.1. ESM-PG items

PG reactions in daily life were assessed with 11 items that map onto the criteria of the DSM-5-TR for PGD and were developed using cognitive interviewing with ESM and/or grief experts (see Table 3). The 11 ESM-PG items correspond to the 10 DSM-5-TR PGD symptoms; i.e. one DSM-5-TR PGD symptom (i.e. ‘Intense emotional pain (e.g. anger, bitterness, sorrow) related to the death’) was captured with two ESM items (i.e. sadness and anger). Details about development of ESM-PG items are described elsewhere (Lenferink et al., 2022b). Participants rated to what extent they experienced each of the 11 DSM-5-TR PGD reactions in the past three hours on a 7-point Likert scale from 0 (not at all) to 6 (very much). These items have sound psychometric properties (Ergun et al., 2025).

Table 3.

ESM-PG items.

Item Number	Item Name	Item
1	Yearning	In the past three hours, I found myself yearning for him/her
2	Preoccupation	In the past three hours, I had intrusive thoughts or images related to the person who died
3	Identity disruption	In the past three hours, it felt as if a part of me has died along with the deceased
4	Marked sense of disbelief	In the past three hours, It felt unreal that he/she is dead
5	Avoidance	In the past three hours, I avoided places, objects, or thoughts that reminded me that he/she is dead
6	Sadness	In the past three hours, I felt sad because of his/her death
7	Anger	In the past three hours, I felt bitterness or anger because of his/her death
8	Difficulty reintegrating	In the past three hours, It was difficult for me to do something (e.g. social activities, studying, working, sports, hobbies) because of his/her death
9	Emotional numbness	In the past three hours, I felt emotionally numb because of his/her death
10	Meaninglessness	In the past three hours, I felt that life is unfulfilling or meaningless without him/her
11	Loneliness	In the past three hours, I felt alone or detached from other individuals because of his/her death

Note. ESM = Experience sampling methodology; PG = Prolonged grief.

1.5.2. ESM contextual factors

Six items were included to measure the contextual factors during the past three hours in all three projects. Namely, physical environment (i.e. home, school, work, other), current activity (e.g. resting, household chores, sports), quality of the activity (i.e. from 0 (very unpleasant) to 6 (very pleasant)), social environment (i.e. alone, with one other person, with multiple others), if not alone, type of relationship with the other person (i.e. romantic partner, child, parent, sibling, friend, acquaintance, colleague/fellow student, other), and quality of the social contact (i.e. from 0 (very unpleasant) to 6 (very pleasant)). Project 3 also included question regarding the pleasantness of being alone (i.e. from 0 (very unpleasant) to 6 (very pleasant)). You can find the detailed item information under https://doi.org/10.17026/SS/LNFAXI (File name: 02 Codebook Grief-ID.ods).

1.5.3. Additional ESM-items

Project 3 included six additional ESM questions. First, quality of sleep that was rated in the first assessment of the day with the question ‘I slept well last night’ derived from Forkmann et al. (2018). Second, we developed a single item to assess general feeling of grief (‘In the past three hours, I have experienced grief’). Third, rumination was measured using two items regarding thinking about problems (‘In the past 3 h, I was thinking about my problems’) and thinking about feelings (‘In the past 3 h, I was thinking about my feelings’) derived from Moberly and Watkins (2008). Fourth, the construct of blame was measured with two items regarding blaming others for the loss (‘In the past 3 h, I felt intense blame toward others because of his/her death’) and blaming self for the loss (‘In the past 3 h, I felt intense blame toward myself because of his/her death’) derived from TGI-SR + (Lenferink et al., 2022a) item measuring blame.

2. Data description

Table 4 details the archive sample composition. In total, 315 people are included in this archive, with 22,050 ESM measurement points in total. Majority of the participants were middle-aged, identified as woman, and completed higher education. Almost half of the sample experienced the loss of a close person namely partner or a child, majority of the losses were due to natural causes, and happened less than a year ago. On average, participants responded to 43.77 ESM notifications in total, with the median being 50, on a scale from 1 to 70, which reflects an overall compliance rate of 71%

Table 4.

Participant characteristics (N = 315).

Characteristic
Age group, N (%)
Early adulthood (18-44)	89 (28.34%)
Middle adulthood (45-64)	170 (54.14%)
Late adulthood (>65)	55 (17.52%)
Gender, N (%)
Female	256 (81.53%)
Male	58 (18.47%)
Other	0 (0%)
Education, N (%)
Higher education	189 (60.19%)
Other education (primary school, high school or vocational education)	125 (39.81%)
Cause of death, N (%)
Natural loss (physical illness, other)	220 (70.06%)
Traumatic loss (accident, suicide, homicide)	94 (29.94%)
Kinship (the deceased was my …), N (%)
Partner, child	153 (48.73%)
Parent, sibling, grandparent, grandchild, friend, other	161 (51.27%)
Time Since Loss, N (%)
≤ 6 months	155 (49.21%)
> 6 months and < 12 months	30 (9.52%)
≥ 12 months	130 (38.10%)
Number of completed ESM notifications, M (SD), median	43.77 (20.45), 50

Note. ESM = Experience sampling methodology.

2.1. How is it FAIR?

2.1.1. Findable

The archive is findable via the open source repository DANS. The metadata of this archive is findable by anyone without logging into DANS, under the link: https://doi.org/10.17026/SS/LNFAXI. The metadata description includes keywords, data owners, methods description, and other information, thereby facilitating direct access to the metadata. The DANS data station ensures that the dataset is machine-readable by creating a machine-readable metadata format (i.e. json). The uploaded archive dataset is referenced with a unique and persistent identifier (i.e. doi: 10.17026/SS/LNFAXI). Furthermore, each dataset included in the archive is assigned a unique persistent identifier, which is provided in the archive metadata. This repository provides a secure environment for the storage of data in accordance with the General Data Protection Regulation (GDPR) for Europe, thereby ensuring the protection of personal information.

2.1.2. Accessible

Data can be accessed through the persistent identifier or directly through the DANS repository. The repository contains metadata files with detailed characteristics of the datasets, which are openly shared. The archive dataset, however, is not fully open, but is available under restricted access to qualified researchers. The dataset can be accessed by clicking on ‘Request Access’ in DANS for a specific file that the user wishes to receive. To do this, users must have an account in DANS and log in to submit this request. For the request to be approved, the investigators should provide their name, contact details, affiliation, and a link to their preregistration of their research project, see Mertens and Krypotos (2019) for more information on preregistering secondary analyses on existing data. Data contributors (i.e. people who have contributed data to this archive) evaluate the request based on whether (a) the research questions can be addressed with the data requested, (b) appropriate data security measures will be taken, (c) the investigators agree to acknowledge the source in presentations and publications (see below for reuse instructions), and (d) the investigators agree not to share the data with third parties. Supplementary Material B includes the example data request form. A record of the projects that reused the archive data (i.e. a list of shared preregistration links) will be made available to other researchers in the archive metadata to foster further collaborations and comply with open science principles. The authors of this archive are open to collaborations on a variety of research questions, although this is not a requirement.

2.1.3. Interoperable

The data are clearly defined, with labels and different types of files available for other users using different open source programmes. Open specifications are not dependent on any particular software, developer, or vendor. The files are presented in formats that ensure long-term sustainability and accessibility. The preferred formats are PDF/A for text documents, CSV or ODS for spreadsheets and databases, and numeric data in DAT, CSV, and R formats. Other non-preferred formats, which are nevertheless relevant to the field, are also available (e.g. SAV for use with SPSS software), but their long-term sustainability and availability are not ensured.

2.1.4. Reusable

The files and variables are explained in detail, including different user-readable metadata to facilitate the reuse of this archive. The Readme file shows how the files are organized. It includes a data dictionary and a codebook that describe the variables, sample composition, sampling procedure, response/non-response rates, and data collection method. When reusing the archive, the researchers are responsible for providing appropriate acknowledgements, including citations for the archive, this paper, and the corresponding individual datasets that were used for the analyses. The investigators are instructed to store the data securely and are instructed not to share the data with others.

Researchers are welcome to contribute their data to this archive, for this, the instructions (e.g. harmonization and data wrangling R codes) and minimum requirements are provided. These requirements include: (a) data were collected in a study that has received ethical approval from an institutional review board; (b) the authors have the right to share the data with other researchers; (c) the ESM-data are collected at least once daily for at least seven consecutive days; (d) there should be sufficient overlap with the measured constructs included in the archive.

3. Discussion

The development of this archive demonstrates that it is feasible to combine research on PG reactions in their natural environment, ensure that it is FAIR and make it available for reuse. The archive currently consists of three datasets from three projects, 315 participants and over 20,000 measurement points. To date, no other data archive contains information about PG reactions multiple times a day over an extended period of time, therefore, the archive presents new opportunities for data analyses addressing unresolved research gaps.

The utility of this archive is evidenced as its being used in collaborative and international initiatives. To illustrate, the archive has been used to investigate fluctuations in two symptoms: yearning and sadness (Marcolini et al., 2025), with five distinct fluctuation profiles emerging. The archive has been utilized to examine the associations between social context and PG symptoms among traumatically bereaved people. Findings indicate that when people enjoyed their social life more than usual, they experienced lower levels of grief. However, differences in social life quality compared to peers were not related to PGD levels (Specker et al., 2025). In addition, the archive made it possible to study networks of PG symptoms over time. The results showed that when a person experiences emotional numbness, it is likely that other symptoms will increase about three hours later (Pociūnaitė-Ott et al., 2025).

3.1. Limitations and future directions

There are a few limitations that should be acknowledged to guide future research efforts. First, the data presented in this archive are specific to a Dutch sample, with a small proportion of German participants. This should be considered when generalizing the findings to non-Dutch people. Second, the projects harmonized in this archive include samples with varying compositions, specifically differences in the time since loss or type of loss. Although this may be perceived as a limitation, it also contributes to greater heterogeneity of the data, thereby facilitating answering questions that rely on heterogeneous data (e.g. subgroup-analyses based on type of loss). Third, the archive includes missing data, which is inevitable due to the intense longitudinal data collection in natural settings (Fritz et al., 2024). In such cases, available techniques for handling missing data should be used.

Future developments of this archive will necessitate some harmonization efforts. Given the lack of standardized measures in ESM research, we anticipate a considerable degree of cross-study variation. Harmonizing ESM datasets poses several unique challenges, such as discrepancies in sampling frequency (e.g. studies collecting data five versus seven times per day), differences in item wording, or timing of measurements. In such cases, potential solutions may include aggregating items to a once-daily level, aligning overlapping timepoints, or treating unmatched timepoints as missing data. Gathering expert input and drawing on harmonization strategies from other fields (Kassam-Adams et al., 2020; Pociunaite et al., 2023) will be particularly helpful in addressing these challenges.

3.2. Conclusion

This archive provides a valuable dataset for investigating research gaps related to daily grief reactions and for disentangling between-person and within-person effects. Given the resource-intensive nature of data collection in ESM research, we encourage other researchers to reuse the data from this archive, thereby furthering our understanding of grief in natural environments. Moreover, researchers who have studied grief in daily life are warmly encouraged to contribute their data to the archive. As a living resource, the archive is designed to continuously grow, and all contributors will receive appropriate recognition for their involvement. By presenting our approach to making the archive FAIR, we aim to inspire others to adopt similar practice. As the archive continues to grow we expect the scientific discovery of grief research to progress more rapidly and extensively. Further information about the archive is available on the website www.griefresearch.nl/grief-id/, including instructions for requesting or contributing data.

Funding Statement

Lonneke I. M. Lenferink is funded by the project ‘Toward personalized bereavement care: Examining individual differences in response to grief treatment’ [ID: Vl.Veni.211G.065] of the research programme [NWO Talent Programme 2021 – Veni] which is financed by the Dutch Research Council (NWO). Justina Pociūnaitė-Ott is funded by Trauma Data Institute.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Archive data and metadata are available at Data Archiving and Networked Services (DANS) repository via this link .

Open Scholarship

This article has earned the Center for Open Science badges for Open Data and Open Materials through Open Practices Disclosure. The data and materials are openly accessible at https://doi.org/10.17026/SS/LNFAXI and https://doi.org/10.17026/SS/LNFAXI

Supplemental Material

Supplemental map for this article can be accessed at https://doi.org/10.1080/20008066.2025.2526885